If you asked me to design a traffic light I wouldn’t take the job unless I had years to get up to speed. I drive, I use traffic lights every day, but I don’t have a deep understanding of how they work or their context. Thus, I am ill equipped to design a traffic light well.

The dynamics of traffic flow, connectivity with other systems, the ability for law enforcement to override the signals… I don’t have any idea how any of that works, let alone how to improve it. And that’s just the stuff I have a vague awareness of! There are doubtless significant aspects of traffic light design that I don’t even know about. I don’t know what I don’t know, and that’s an incredible obstacle to designing well.

I could attempt to create a working traffic light right now, but it would be full of flaws. Issues would arise the instant I hit the limit of my current knowledge. I would be forced to make assumptions, and every assumption would be another chance to do things horribly wrong. Bad design is often the product of assumptions, and my traffic light would be full of them.

If you want to design something well you need to start with a deep understanding of the problem you’re trying to solve, the goals you’re trying to accomplish, and the context of it all. Without that, you’re just guessing and rolling the dice.